Despite the advent of potent anti-retroviral therapy (ARV) and extensive efforts characterizing the mechanisms of HIV disease, unresolved, complex challenges in HIV research remain that demand a multidisciplinary approach integrating the basic and clinical sciences. In this spirit, the premise of this application is that mathematical nonlinear dynamical system models of within-host HIV dynamics coupled with statistical population models, integrated with clinical and biological expertise and informative data can accelerate breakthroughs in HIV research. A unique, multidisciplinary team merging expertise in immunology and clinical investigation with expertise in mathematical and statistical modeling, whose record of fruitful collaboration is already established, will carry out joint research on mathematical, statistical, immunological, and clinical developments toward this common goal through five interwoven specific aims. Although approximations, mathematical models of within-subject HIV dynamics can yield insights into mechanisms underlying HIV pathogenesis. A model developed in the last project period yields accurate long- term predictions of individual subject longitudinal immunologic and virologic profiles. The first aim involves extending the model to incorporate more realistic representation of body's immune response to the virus, enhancing its predictive ability. Applying the models to data in this way requires an appropriate statistical framework and mastery of the accompanying computational challenges. The second aim is to develop and implement practical statistical methods that can address these challenges and be used to gain information on dynamics in the population, so that the models can be used as the basis for the next aim. The predictive capability of these models when applied to data suggests that they can be a powerful tool in the design of clinical trials. The third aim involves development of a systematic strategy for using model-based simulation to inform the design and conduct of clinical trials, which has the potential to lead to more time- and cost- efficient HIV clinical research. Both to prove this principle and to address a key, outstanding clinical question, the fourth aim is to use this approach to design and conduct a clinical trial to determine whether treatment initiated during acute infection followed by terminal interruption at time(s) determined by the models, results in a lower viral load set point and higher CD4 cell count than no treatment. Finally, the fifth aim is to develop and use optimal control theory, mathematical theory for modifying the behavior of nonlinear dynamical systems through control of system inputs, to suggest new, practical adaptive HIV treatment strategies that may lead to improved long-term outcomes. The rich longitudinal data collected during the trial will be used to facilitate development of these strategies.